When list-watch suffer from network , some kind of warning or alert may be helpful

[panpan0000]

What would you like to be added?

If there's a CRD or a prometheus metrics which tell user the sync is outdate , it will be a direct signal and helpful to drive the user to diagnostic the network problem directly,

Why is this needed?

For example, when the outbound network bandwidth is full (especially the public cloud), or when the network condition is rather bad, the list-watch will be out of sync , and the data will be out of date as well.

If there's a CRD or a prometheus metrics which tell user the sync is outdate , it will be a direct signal and helpful to drive the user to diagnostic the network problem directly,

[clusterpedia-bot]

Hi @panpan0000, Thanks for opening an issue! We will look into it as soon as possible.

Details

Instructions for interacting with me using comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the gh-ci-bot repository.

[Iceber]

This log just shows that the list took a long time, but eventually the list succeeded and started watching

pediacluster has a synchronization status field for the resource, which shows specific information when the resource synchronization stops or is abnormal

status:
...
  syncResources:
  - group: apps
    resources:
    - kind: Deployment
      name: deployments
      namespaced: true
      syncConditions:
      - lastTransitionTime: "2023-01-10T09:46:55Z"
        status: Stop
        storageVersion: v1
        version: v1

We can add another state between list and watch, Syncing state means when the list finishes and watch starts, we can use Listing or other states when starting the list